Timezone: »

 
What Matters in Learning from Offline Human Demonstrations for Robot Manipulation
Ajay Mandlekar · Danfei Xu · Josiah Wong · Chen Wang · Li Fei-Fei · Silvio Savarese · Yuke Zhu · Roberto Martín-Martín

@
in Offline Reinforcement Learning »

Imitating human demonstrations is a promising approach to endow robots with various manipulation capabilities. While recent advances have been made in imitation learning and batch (offline) reinforcement learning, a lack of open-source human datasets and reproducible learning methods make assessing the state of the field difficult. In this paper, we conduct an extensive study of six offline learning algorithms for robot manipulation on five simulated and three real-world multi-stage manipulation tasks of varying complexity, and with datasets of varying quality. Our study analyzes the most critical challenges when learning from offline human data for manipulation. Based on the study, we derive a series of lessons including the sensitivity to different algorithmic design choices, the dependence on the quality of the demonstrations, and the variability based on the stopping criteria due to the different objectives in training and evaluation. We also highlight opportunities for learning from human datasets, such as the ability to learn proficient policies on challenging, multi-stage tasks beyond the scope of current reinforcement learning methods, and the ability to easily scale to natural, real-world manipulation scenarios where only raw sensory signals are available. Upon acceptance, we will open-source our datasets and all algorithm implementations to facilitate future research and fair comparisons in learning from human demonstration data. Additional results and videos at https://sites.google.com/view/offline-demo-study

Author Information

Ajay Mandlekar (Stanford University)
Danfei Xu (Stanford University)
Josiah Wong (Stanford University)
Chen Wang (Stanford University)
Li Fei-Fei (Stanford University)
Silvio Savarese (Stanford University)
Yuke Zhu (University of Texas - Austin)
Roberto Martín-Martín (Stanford University)

More from the Same Authors